Term 2 Week 5: Time Preferences and Sin Taxes Flashcards
What is the economic rationale for consumption taxes (3)
-Correct for harmful effects of consumption on third parties (pigovian taxes)
-Raise government revenue
-Reduce consumption of G/S with potential danger to health/wellbeing
What is discounting and why does it occur (3,3)
-Decisions taken over time involve discounting
-We typically assume a constant discount rate δ
-if δ = 0.9, £100 next year is worth £90 etc.
Money in the future is discounted relative to today, as:
-Opportunity cost
-Patience
-Probability we actually get to period t
What is present bias and how does hyperbolic discounting account for this (2,3)
-Present bias is when people tend to overvalue immediate rewards relative to all future rewards
-People have a preference to incur positive payoffs today, and postpone negative payoffs, only to reverse that preference when the future approaches
-Standard discounting measures total utility as U = u0 + δu1 + δ^2u2 …
-(Quasi)Hyperbolic discounting measures total utility as: U = u0 + δβu1 + δ^2βu2 …
-β < 1 indicates a present bias, β = 1 is no present bias
What is the main government justification of sin taxes (1)
-One government justification of implementing a sin tax (alcohol, sugar etc) is that consumers have present bias, and will prioritise the benefits of consumption now over the health cost later
How can we derive the implementation of sin taxes on the consumer maximisation problem (2,3,2)
-Consumers have discounted utility U = u0 + βδu1 + βδ^2u2 + …
-Here, β < 1 implies a present bias
-Say there are 3 ways for consumers to spend their budget (B), x = crisps, y = carrots, z = money
-At each time period the consumer solves maximisation problem: ln(x) + ln(y) + z - βδγln(x) s.t. B = (px+tx)x + (py + ty)y + z
-γ<1 measures the harm done in the next period of consuming crisps
The consumer maximisation problem will give demand for x:
-x* = (1-βδγ)/(px + tx)
-y* = 1/(py + ty)
How can we derive the implementation of sin taxes with government action (1,3,2,3)
-First of all, normalise px = py = 1
-As a benevolent government, we maximise over tx, ty
-Ū = ln(x) + ln(y) + z* - δγln(x) s.t. txx + tyy* = R
-Since the government is not present biased, they don’t have the β factor
Using the lagrangian, -1 + tx* = μ(1-βδγ)/(1-δγ)
-1 + ty* = μ
-This tells us tx - ty > 0 when (1-βδγ)/(1-δγ) > 1
-This happens iff β<1
-Hence, a present bias necessitates a larger tax on the sin good
What is the slope of the indifference curve of goods x and y (gov vs consumer), and what does this tell us about their optimal consumption (2,1)
-The government’s IC slope = -MRS = -(1-δγ)y/x
-The government’s IC slope = -MRS = -(1-βδγ)y/x
-When plotting these both against a budget constraint, the governments optimal consumption point is less x and more y than the consumer
What is the difference between sophisticates and Naifs (2,2)
-If some consumers had present bias and some didn’t, there will be a tradeoff between β < 1 and hurting β = 1
-The effect is larger on β < 1 types, as demand is higher
-Sophisticates would vote for this tax, understanding they have a lack of control
-Naifs would believe they would be fully self controlled in the future
What are the pros and cons of sin taxes (2,4)
+Can combat serious public health problems
+Generates revenue by taxing inelastic products
-Could be regressive (UK sugar tax annual CV is 40% higher for bottom half of income distribution)
-Could encourage substitutes consumption
-Could encourage the black market
-Paternalistic (is it the governments place to intervene)